* Developed new dashboards and analytics web-app tools.* Designed new Javascript-based dashboard tools integrated with a Druid datastore to provide highly responsive tools with high cardinality datasets spanning years of streaming history. Created Druid datasets to power these dashboards with tDigest, HyperLogLog, and arithmatic aggregations.* Developed new metrics to characterize delivered quality of experience to Netflix members across the globe. * Created predictive models to pre-tag triggered device quality alerts in order to reduce the number of false positives reported to our Device Reliability Team. This alert volume reduction resulted in savings equivalent to 0.5 FTE person-hours of effort per year in triaging noisy alerts.* Create an alerting framework to monitor partner firmware rollouts and monitor device quality metric movement correlated with the new firmware. With this framework, we identified issues with buggy firmware deployments from our consumer electronics partners which would have impacted millions of Netflix customers while only deployed to ~10,000 devices and work with our partners to minimize impact.* Developed forecasting models to help engineering teams understand the potential device ecosystem 2-3 years into the future including the mixture of Netflix SDK version breakdown.* Created a framework to identify common dimensions among many rows in a dataframe of "interesting" events. Used market basket analysis to find sets of dimensions to occur at increased rates among these events.* Worked with device ecosystem teams to develop a new metric to monitor device availability to help quantify member-facing device quality issues and their impact on member joy.* Modeled device hardware's impact on in-field device quality, including creating model-produced proxies for hardware parameters not directly observable. Created device segmentation to aid in AB testing of treatment to understand the impact to "high-end" and "low-end" devices.

* Created numerous dashboards (in Tableau and Javascript) and reports (in Python, and Jupyter Notebooks) to surface metrics around member usage, activation, retention, engagement, and device performance across many consumer electronic device types to inform business decisions and identify anomalous trends to minimize customer impact.* Provided insights around Netflix Ready Device Platform feature reach, usage, and efficiency as well as quantifying the capabilities of Netflix-enabled devices to shape development roadmaps and priorities.* Designed classification tools to enable engineering teams by triaging daily alert emails to prioritize those likely of being the most critical.* Performed launch analysis and after action reports to provide real-time insight for major partner launches and to understand the user impact of device performance issues in the field.* Developed new metrics and ETL to identify specific device behavior fingerprints associated with user-facing errors with no associated logging in order to aid engineering teams.* Interfaced with engineering teams to ensure needed logging is implemented and aided in testing prototyped changes.

• Created automated process to analyze the 10-100 gigabytes of imaging taken each night at the observatory and log data quality metrics.• Implemented k-mean clustering analysis to monitor system performance which avoided 4+ critical failures due to failing equipment.• Developed a logistical regression model to inform night-time operators procedures, resulting in a 20% reduction in overheads between observations.• Created a queue scheduling system to maximize time efficiency of night-time observations.• Used logistic regression techniques to identify most important features contributing to atmospheric turbulence at the observatory.• Maintained and upgraded observatory software to provide more efficient user experience• Trained astronomers in proper observing procedures. • Studied user feedback to prioritize proposed software improvement and development.• Supervised queue observers and advised them in their nightly decision making.

• Mined the largest available public astronomical databases using SQL • Performed exploratory analyses of data sets to identify trends and interesting features• Quantified statistical properties of data sets as a probe of galaxy evolution• Applied machine learning algorithms including regression, classification, and clustering (k-mean and kth nearest neighbor) to millions of galaxies to study the distribution function of galaxies properties, their interdependence, and their evolution• Lead team tasked with processing terabytes of astronomical data, extracting tractable datasets, and deriving statistical measurements • Implemented Bayesian and p-value hypothesis testing techniques as well ass A/B testing to constrain models for galaxy evolutionary processes. • Created data archive and software repository for survey products • Strong record of peer-reviewed publications.

• Statistical Analysis of millions of galaxies • Designed experiments to provide data utilized in Bayesian hypothesis testing • Lead random forest classification effort of millions of galaxies • Lead maximum likelihood analysis of evolving properties of galaxies

• Performed research activities in a broad range of astrophysical fields. • Taught undergraduate students introductory topics in astronomy.• Communicated technical principles to a non-technical audience.

Performed guided research centering on the morphology and variability of active galaxies with the VLBA.